The human capital intervention evaluations undertaken in Projects I through III will provide an assortment of impact estimates of interventions on achievement, anti-social behavior and, in the case of Project II, health risk behavior. The current project would address the ultimate so what? question of the possible longer-run consequences of augmenting these skills and improving behavior in middle childhood and in adolescence. Specifically, we assess to what extent math and reading achievement and anti-social behavior in middle childhood are predictive of adult labor market outcomes and of avoiding serious adult crime as well as the role of health risk teen behaviors on adult health. Data are drawn from six population-based developmental data sets (the Swedish Study of Individual Development and Adaptation, the Finnish Jyvaskyla Longitudinal Study of Personality and Social Development, the British Cohort Study-1970 birth cohort and the British National Child Development Survey-1958 birth cohort, the U.S. Baltimore Beginning School Study, and the child sample from the U.S. National Longitudinal Survey of Youth), virtually all of which measure reading and math skills and anti-social behaviors between ages 7 and 10, and again in adolescence, and follow their samples of children into middle adulthood (in our data sets, at least age 25 and, in two cases, as late as age 50). We use these longitudinal data sets to: i) describe cross-country differences in age 7-10 skills and behaviors and age 25-50 outcomes by various measures of parental socioeconomic status; ii) estimate associations between middle-childhood skills and adult labor market, crime and health outcomes for each of our samples and identify social-class differences in these associations; iii) assess the extent to which adolescent skills and behaviors mediate the explanatory power of middle childhood skills and behaviors in predicting adult outcomes; and: iv) in the case of middle-childhood skills and adult health, estimate to what extent teen health risk behaviors account for the links. Within the limits of its nonexperimental analyses, this project's search for convergent findings across populations and countries complements the other three in the program project by showing which skills and childhood stages targeted by the various education-related interventions we evaluate matter most for adult well-being.